Data Mining Optimization Based on Particle Swarm Optimization For Diagnosis of Inflammatory Liver Disease

Abstract

Inflammation of the liver is a contagious disease that is a public health problem that affects morbidity, mortality, public health status, life expectancy, and other socio-economic impacts. Early diagnosis of this disease is very important so that it can be quickly treated and treated. In this study the researchers will apply and compare several data mining and optimization classification methods with particle swarm optimization (pso), including the C4.5 algorithm, k-Nearest Neighbor, C4.5 with PSO, and k-Nearest Neighbor with PSO to diagnose inflammatory diseases. carefully, then compare which of the several of these methods is the most accurate. Based on the results of measuring the performance of the three models using the Cross Validation, Confusion Matrix and ROC Curve methods. Based on the research results, it is known that the C4.5 method with PSO is the best method with an accuracy of 79.51% and an under the curva (AUC) value of 0.950, then the k-Nearest Neighbor method with PSO has an accuracy of 75.59% and an AUC value of 0.909, then the C4.5 method with an accuracy rate of 70.99% and an AUC value of 0.950, then the k-Nearest Neighbor method with an accuracy rate of 67.19%, and an AUC value of 0.873. This proves that particle swarm optimization can improve the performance of the classification method used.